Natural Language Processing news and analysis on Web Pulse: 5580 curated articles tracking the latest Natural Language Processing developments, tools, and research, updated continuously from vetted sources.
Researchers developed a two-stage system using large language models to extract fine-grained event tags from SEC 8-K filings, mapping disclosures to a 119-event taxonomy with grounded quotes. Applied to 292,984 filings f…
The ICDAR 2026 HIPE-OCRepair competition evaluated LLM-assisted OCR post-correction for historical documents in English, French, and German. Four teams submitted systems using strategies from zero-shot prompting to fine-…
Researchers propose a multi-cluster boundary learning method for out-of-scope intent detection using MiniLM embedding, achieving state-of-the-art performance on CLINC150, StackOverflow, and Banking77 datasets. The method…
A new survey examines the alignment of large language models with clinical reasoning needs, proposing a five-level competency scheme based on Miller's Pyramid and evaluating 18 state-of-the-art models on a benchmark data…
Researchers released the Nigeria Machinery Usage and Failures Dataset, containing 89 machine-level records across 28 indicators from Nigeria's manufacturing and oil and gas sectors (2006–2025), along with a chain-of-thou…
Researchers released Infinity-Parser2, a large multimodal model for end-to-end document parsing, along with a 5-million-sample bilingual corpus Infinity-Doc2-5M. The model uses multi-task reinforcement learning across ei…
InternNav, an open-source toolbox for embodied navigation built on PyTorch, Habitat, and Isaac Sim, has been released. It supports modular navigation systems, multiple simulation platforms, and includes the InternVLA-N1 …
AudarAI released Audar-ASR-V1, an open-weight speech recognition model focused on Arabic and its dialects, with plans to expand to multilingual audio intelligence. The release includes Flash and Turbo models, a technical…
A developer at Nylas built an SDR agent that uses a Nylas Agent Account as its own mailbox, enabling it to send sequences, receive replies, classify them, and stop the sequence when a real reply arrives. The agent accoun…
OpenAI released gpt-realtime-2.1 and gpt-realtime-2.1-mini voice models on July 6, featuring a 25% reduction in p95 latency through caching improvements and adding reasoning and tool use to the mini model. The update eli…
AI is transforming email marketing workflows by compressing days of manual work into hours, shifting marketers from creators to editors. Instead of replacing humans, AI handles repetitive tasks like research, drafting, a…
Granola, an AI-powered meeting note taker, reduces manual effort by capturing device audio locally and using AI to enhance rough bullet points into structured notes, addressing the cognitive costs of manual note-taking a…
AI meeting tools can now listen to conversations and generate structured notes, but the method of capture affects participant trust. Granola offers a privacy-first approach by capturing audio locally without a visible bo…
Executive recruiters are adopting AI note-taking tools like Granola that capture interviews without visible recording bots, preserving candid conversations. The tool runs locally on the user's device and enhances rough n…
Apple's iOS 27 introduces over eight new CarPlay features, including video browsing while parked, Siri AI for hands-free conversations, a persistent mini-player, new wallpapers, route data sharing with vehicles, improved…
Hugging Face's smolagents library enables AI agents to generate and execute SQL queries against databases, using a code-based reasoning approach that allows error correction and verification. The agent writes Python snip…
Qdrant reduced retrieval-augmented generation token costs by 67% using native ColBERT reranking, which performs token-to-token matrix comparison inside the database in a single query call, eliminating the need for extern…
Databricks held its fifth Genie Hackathon, where teams built projects using Databricks Genie products that let users work with data in plain English. OneTrust built a supervisory layer to shard data across multiple Genie…
Gartner's July 7, 2026 Magic Quadrant for Conversational AI Platforms elevated SoundHound AI and Salesforce to the leaders quadrant, while IBM moved to visionaries and Netomi debuted as a challenger. The report highlight…
Orbit, an open-source AI gateway for retrieval-based inference, has been released, enabling self-hosted private RAG, natural-language data access, and tool-calling agents across 37+ model providers. The toolkit supports …